We came upon an interesting post at JVA Consulting’s blog related to an important topic in the nonprofit world- is our work accomplishing something or are we just doing good. We’ve worked at places where the emphasis was on ‘good works’ rather than good results, so we appreciate how important this is. The post talked about the value of logic models as ‘road maps’ for laying out how program activities are linked to short and longer term outcomes. Logic models certainly can be seen as road maps but for us their value is really mostly in the process of creating them. In this sense they are heuristic devices that activate the process through which stakeholders make explicit the ideas and ideologies that underlie their work. You can find a tutorial we did on logic models here.
We believe the best way to go about creating a logic model is to start with the desired outcomes of a program. The logic model tracks backwards through the steps needed to bring it about, allowing consideration for both the material and philosophical assumptions that underlie the change. The presence of resources would be an example of material assumptions (e.g., management will be available to supervise the new program). A belief that people will change their behavior based on new knowledge they acquire is an example of a philosophical assumption (a shaky one at times, if you ask us).
In our experience, logic models can sometimes be pro forma exercises in which stakeholders build lists of assumptions, activities, short term and long term outcomes which they then connect with a spaghetti of lines and links that obfuscate more than clarify. The theory of change technique provides a much more structured methodology for creating logic models. It eschews the rigid structure of the typical logic model and compels those using it to articulate the many short and intermediate outcomes that need to occur before the initiative’s ultimate goals can be achieved. Each outcome is specified with an indicator and an associated threshold value above which the outcome can be said to have been achieved. The venture’s ultimate outcomes are brought about by interventions represented by the lines connecting several levels of intermediate outcomes. This is a key difference between logic modeling and the theory of change approach. Graphically, it does not seem to mean that much– interventions are now lines whereas in a logic model they are listed in a column headed ‘activities’. Conceptually however, it means a lot. The theory of change approach emphasizes the milestones on the way to achieving a program’s goals and more tightly integrates measurement and assessment into the program design. Just as important, it allows for many more intermediate outcomes than one typically finds in a logic model. While logic modeling moves the emphasis from program activities to program outcomes, the Theory of Change technique takes this a step further by placing them at center stage. The Aspen Institute has a great primer on the Theory of Change approach which you’ll find here. We plan to use it in place of logic modeling in our work in the future.
Still, we believe that logic models and theory of change diagrams can take over program and evaluation planning. Wisely, the Aspen Guide recommends limiting intermediate outcomes to five levels. However, we have seen theory of change diagrams that have many more levels than that and the same spaghetti of lines that become, in short order impossible to follow. A power point presentation comparing logic modeling and the Theory of Change technique can be found here. We’re planning on creating a white paper on the subject in the near future. If you’d like to see it, please drop us a line and we’ll send you a draft when we’ve got it done.